Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 25687
    Prev
    Next
    INDEX
    Explanations

    terms related to political processes and their implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    288
    +0.12
    0.5%
    513
    +0.10
    0.5%
    216
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.06
    513
    +0.10
    0.03
    170
    +0.10
    0.03
    Negative Logits
    sites
    -0.77
     Blocks
    -0.73
     IDs
    -0.73
    ainers
    -0.72
    Us
    -0.70
    points
    -0.69
     launchers
    -0.68
    lations
    -0.68
    cases
    -0.67
     channels
    -0.67
    POSITIVE LOGITS
     dominated
    1.03
     rife
    0.95
     ravaged
    0.92
     riddled
    0.90
     plagued
    0.90
     fraught
    0.88
    ustomed
    0.85
     devoid
    0.85
     starved
    0.85
     accustomed
    0.84
    Activations Density 0.184%

    No Known Activations